A meteorological and chemical overview of the DACCIWA field campaign in West Africa in June–July 2016

HAL Id: insu-01518455

https://hal-insu.archives-ouvertes.fr/insu-01518455

Submitted on 4 May 2017

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

DACCIWA field campaign in West Africa in June–July 2016

Peter Knippertz, Andreas H. Fink, Adrien Deroubaix, Eleanor Morris, Flore Tocquer, Mat J. Evans, Cyrille Flamant, Marco Gaetani, Christophe

Lavaysse, Céline Mari, et al.

To cite this version:

Peter Knippertz, Andreas H. Fink, Adrien Deroubaix, Eleanor Morris, Flore Tocquer, et al.. A

meteorological and chemical overview of the DACCIWA field campaign in West Africa in June–July

2016. Atmospheric Chemistry and Physics, European Geosciences Union, 2017, 17 (17), pp.10893-

10918. �10.5194/acp-17-10893-2017�. �insu-01518455�

https://doi.org/10.5194/acp-17-10893-2017

© Author(s) 2017. This work is distributed under the Creative Commons Attribution 3.0 License.

A meteorological and chemical overview of the DACCIWA field campaign in West Africa in June–July 2016

Peter Knippertz

, Andreas H. Fink

, Adrien Deroubaix

, Eleanor Morris

, Flore Tocquer

, Mat J. Evans

, Cyrille Flamant

, Marco Gaetani

, Christophe Lavaysse

, Celine Mari

, John H. Marsham

, Rémi Meynadier

, Abalo Affo-Dogo

, Titike Bahaga

, Fabien Brosse

, Konrad Deetz

, Ridha Guebsi

, Issaou Latifou

,

Marlon Maranan

, Philip D. Rosenberg

, and Andreas Schlueter

1

Institute of Meteorology and Climate Research, Karlsruhe Institute of Technology, 76128 Karlsruhe, Germany

2

Laboratoire de Météorologie Dynamique, Ecole Polytechnique, IPSL Research University, Ecole Normale Supérieure, Université Paris-Saclay, Sorbonne Universités, UPMC Univ Paris 06, CNRS, 91128 Palaiseau, France

3

Wolfson Atmospheric Chemistry Laboratories, Department of Chemistry, University of York, York, YO10 5DD, UK

4

Laboratoire d’Aérologie, Université de Toulouse, CNRS, UPS, 31400 Toulouse, France

5

LATMOS/IPSL, Sorbonne Universités, UPMC Univ Paris 06, UVSQ, CNRS, 75252 Paris, France

6

European Commission, Joint Research Centre, Ispra (VA), Italy

7

School of Earth & Environment/National Centre for Atmospheric Science, University of Leeds, Leeds LS2 9JT, UK

8

AXA Group Risk Management Department, Paris, France

9

Direction Générale Météo Nationale, B.P. 1505, Lomé, Togo Correspondence to: Peter Knippertz (peter.knippertz@kit.edu) Received: 13 April 2017 – Discussion started: 4 May 2017

Revised: 26 July 2017 – Accepted: 31 July 2017 – Published: 14 September 2017

Abstract. In June and July 2016 the Dynamics–Aerosol–

Chemistry–Cloud Interactions in West Africa (DACCIWA) project organised a major international field campaign in southern West Africa (SWA) including measurements from three inland ground supersites, urban sites in Cotonou and Abidjan, radiosondes, and three research aircraft. A signifi- cant range of different weather situations were encountered during this period, including the monsoon onset. The purpose of this paper is to characterise the large-scale setting for the campaign as well as synoptic and mesoscale weather systems affecting the study region in the light of existing conceptual ideas, mainly using objective and subjective identification al- gorithms based on (re-)analysis and satellite products. In ad- dition, it is shown how the described synoptic variations in- fluence the atmospheric composition over SWA through ad- vection of mineral dust, biomass burning and urban pollution plumes.

The boreal summer of 2016 was characterised by Pacific La Niña, Atlantic El Niño and warm eastern Mediterranean conditions, whose competing influences on precipitation led to an overall average rainy season. During the relatively

dusty pre-onset Phase 1 (1–21 June 2016), three westward- propagating coherent cyclonic vortices between 4 and 13

N modulated winds and rainfall in the Guinea coastal area. The monsoon onset occurred in connection with a marked extra- tropical trough and cold surge over northern Africa, leading to a breakdown of the Saharan heat low and African easterly jet and a suppression of rainfall. During this period, quasi- stationary low-level vortices associated with the trough trans- formed into more tropical, propagating disturbances resem- bling an African easterly wave (AEW). To the east of this system, moist southerlies penetrated deep into the continent.

The post-onset Phase 2 (22 June–20 July 2016) was char- acterised by a significant increase in low-level cloudiness, unusually dry conditions and strong northeastward disper- sion of urban pollution plumes in SWA as well as rainfall modulation by westward-propagating AEWs in the Sahel.

Around 12–14 July 2016 an interesting and so-far undocu- mented cyclonic–anticyclonic vortex couplet crossed SWA.

The anticyclonic centre had its origin in the Southern Hemi-

sphere and transported unusually dry air filled with aged

aerosol into the region. During Phase 3 (21–26 July 2016), a

similar vortex couplet slightly farther north created enhanced westerly moisture transports into SWA and extraordinarily wet conditions, accompanied by a deep penetration of the biomass burning plume from central Africa. Finally, a return to more undisturbed monsoon conditions took place during Phase 4 (27–31 July 2016). The in-depth synoptic analysis reveals that several significant weather systems during the DACCIWA campaign cannot be attributed unequivocally to any of the tropical waves and disturbances described in the literature and thus deserve further study.

1 Introduction

The atmosphere over summertime West Africa is influenced by processes covering a wide range of scales, which can in- teract with each other in complex ways (Lafore et al., 2010;

Redelsperger et al., 2006). The dominating phenomenon is the West African monsoon (WAM), which is mainly driven by the surface pressure contrast between the relatively cool waters of the eastern tropical Atlantic Ocean and the Saha- ran heat low (SHL). The former is related to the installation of the Atlantic cold tongue (ACT) starting in April–May and reaching its maximum horizontal extension in mid-August (Caniaux et al., 2011). At the Equator, colder sea surface temperatures (SSTs) increase the stability of the marine at- mospheric boundary layer and decrease the vertical mixing of momentum, leading to weaker surface southerlies (Wal- lace et al., 1989), while north of the Equator as far as the Guinea coast, the large meridional SST gradient strengthens the surface wind through a hydrostatically induced merid- ional pressure gradient (Lindzen and Nigam, 1987). This cre- ates a low-level circulation characterised by surface wind di- vergence and subsidence over the Equator and convergence and convection close to the Guinea coast in the period before the full onset of the WAM.

The SHL is a lower tropospheric thermal depression in the Sahara desert west of 10

E, which develops in response to the intense surface heating during boreal summer (Lavaysse et al., 2009). The monsoon typically sets in quite abruptly around the end of June accompanied by a shift in the area of main rainfall from the Guinea coast to the Sahel (Fitzpatrick et al., 2015; Sultan and Janicot, 2003). This event is usu- ally preceded by a northward shift in the so-called intertropi- cal discontinuity (ITD), the near-surface confluence zone be- tween southwesterly and northeasterly winds, which marks a northern limit of rainfall occurrence (Fitzpatrick et al., 2016;

Lélé and Lamb, 2010). After the monsoon onset, the mid- tropospheric circulation over West Africa is dominated by the African easterly jet (AEJ), which is caused by the strong meridional temperature and moisture gradient at low levels (Cook, 1999; Wu et al., 2009). The AEJ is maintained by the anticyclonic circulation associated with the monsoonal sub- sidence, which characterises the mid-upper troposphere over

the Sahara (Chen, 2005; Thorncroft and Blackburn, 1999) and above the shallow dry convection in the SHL (Garcia- Carreras et al., 2015; Ryder et al., 2015). Barotropic and baroclinic instabilities associated with the jet create an en- vironment favourable to the generation of African easterly waves (AEWs) (Thorncroft and Hoskins, 1994a, b; Wu et al., 2012), synoptic-scale disturbances characterised by a 2–

6-day period in the Sahel. Diedhiou et al. (1999) present evi- dence for a more intermittent, slower (6–9-day period) wave regime with cyclonic and anticyclonic centres straddling the AEJ, longer wavelengths and an activity maximum over the continent in June and July.

On multi-decadal to inter-annual timescales, WAM vari- ability is strongly associated with global SST anomalies (Rodríguez-Fonseca et al., 2015; Rowell, 2013). For exam- ple, positive phases of the Atlantic multi-decadal variability favour precipitation in the Sahel (Ting et al., 2011; Zhang and Delworth, 2006). On inter-annual timescales, SST variability in the tropical Atlantic modulates the land–sea thermal gradi- ent, leading to meridional displacements of the precipitation belt over West Africa (Losada et al., 2010; Polo et al., 2008).

SSTs over the Mediterranean Sea influence the amount of moisture being transported across the Sahara desert and con- verging over the eastern Sahel (Fontaine et al., 2010; Gae- tani et al., 2010). Inter-annual variability in the WAM is also influenced by the SST variability in the tropical Indian and Pacific oceans, which may trigger stationary waves along the Equator interacting over the Sahel (Mohino et al., 2011;

Rowell, 2001).

On intra-seasonal to synoptic timescales, an important source of variability is the propagation of convectively cou- pled equatorial wave (CCEW) disturbances and the Madden–

Julian oscillation (MJO) (Mohino et al., 2012; Pohl et al., 2009). In addition, variability in the SHL strength and position modulate the distribution of the monsoonal precip- itation in the Sahel in the zonal direction. During periods of a deeper SHL, the shallow cyclonic circulation associ- ated with the thermal low is intensified, strengthening the At- lantic westerly flow and the convergence in the Sahel, which leads to wet (dry) anomalies in the eastern (western) Sahel (Lavaysse et al., 2010b). The SHL phases are modulated on synoptic timescales by both tropical and midlatitude distur- bances (Chauvin et al., 2010; Lavaysse et al., 2010a). Other types of intra-seasonal variability include the Sahelian and quasi-biweekly zonal dipole (QBZD) modes on timescales of 10–25 days (Janicot et al., 2011; Mounier et al., 2008;

Roehrig et al., 2011). Variations in the intensity and position

of the AEJ influence the location, amplitude and propagation

speed of AEWs, which play a crucial role in the modulation

of convective precipitation in the Sahel, mainly through their

influence on thermodynamics and vertical wind shear (Gu et

al., 2004; Skinner and Diffenbaugh, 2013). To the south of

the Sahel, low-level vortices unrelated to AEWs can affect

rainfall (e.g. Fink et al., 2006). Convection in West Africa is

often organised on the mesoscale, particularly in the form of

fast-propagating squall line systems (Fink and Reiner, 2003), but more isolated thunderstorms or showers also occur, e.g.

triggered by the sea-breeze convergence along the Guinea coast (Fink et al., 2010). The moist convection embedded within the monsoon flow has been shown to be intrinsic to the monsoon, and the poor representation of convection in mod- els leads to biases in the WAM (Birch et al., 2014; Garcia- Carreras et al., 2013; Marsham et al., 2013).

The atmospheric composition over southern West Africa (SWA hereafter) during the wet season is a complex combi- nation of air masses transported from remote sources, bring- ing desert dust or biomass burning aerosol, and local anthro- pogenic pollution (Mari et al., 2011). The Sahara desert to the north of the region is the largest aerosol source in the world and the transport of dust southwards is a significant source of aerosol for SWA (e.g. Chiapello, 2014; Shao et al., 2011).

Forest fires in the immediate region are not thought to be sig- nificant in this period, but the transport of biomass burning species from the Southern Hemisphere (SH) has been ob- served (Mari et al., 2008). Anthropogenic emissions from the combustion of fossil fuels, biofuels and refuse are on the rise and expected to keep increasing significantly in the near future due to the rapid growth of cities in the region (Knippertz et al., 2015b; Liousse et al., 2014). Air quality is thus a concern, with multiple sources of anthropogenic emis- sions from domestic open fires, road traffic, street dust, waste burning, oil extraction and refining, ships, industrial activity, power plants, etc. SWA is also characterized by a south–north gradient of vegetation, from rainforest in the coastal belt to the sub-Sahelian savannah in the north. The dense vegeta- tion can emit large quantities of biogenic compounds (iso- prene, monoterpenes, etc.), which profoundly alter the gas and aerosol composition in the region (Mari et al., 2011). The relative role of local biogenic and anthropogenic emissions, the long-range transport of other compounds into the SWA atmosphere, coupled to the peculiar dynamics of the region during the monsoonal period leads to a chemically complex region.

A lack of an observational network adequate to bet- ter understand processes and to evaluate model simula- tions and satellite data has impeded scientific progress in West Africa for a long time and motivated the organisation of large international field campaigns. An early example, which revolutionised the understanding of the WAM sys- tem at that time, is the Global Atmospheric Research Pro- gram (GARP) Atlantic Tropical Experiment (GATE) (Kuet- tner, 1974). The largest such programme in recent decades is the African Monsoon Multidisciplinary Analysis (AMMA), which took place in 2006 with a focus on Sahelian convection (Lebel et al., 2010). More recently, the Dynamics–Aerosol–

Chemistry–Cloud Interactions in West Africa (DACCIWA) project (Knippertz et al., 2015a) organised a major interna- tional field campaign during June and July 2016, focusing for the first time on the most populated southern coastal re- gion of West Africa. In addition to a number of meteorolog-

ical aspects, the DACCIWA campaign also had a focus on atmospheric composition, including questions of air pollu- tion and cloud–aerosol interactions (Knippertz et al., 2015b).

Field activities included measurements from three inland ground supersites (Savé in Benin, Kumasi in Ghana, Ile-Ife in Nigeria), urban sites (Cotonou in Benin, Abidjan in Côte d’Ivoire), radiosondes and three research aircraft stationed in Lomé (Togo). A detailed description of the field activities is given in Flamant et al. (2017).

The objectives of this paper are to (a) place the campaign period June–July 2016 into a larger-scale climatological con- text, (b) describe the behaviour of the WAM system (e.g. on- set, AEJ and SHL positions), (c) characterise the most im- portant synoptic-scale weather systems affecting SWA (e.g.

AEWs, vortices), and (d) discuss impacts on rainfall, clouds and atmospheric composition. This way the paper aims to fulfil a similar role as Janicot et al. (2008) for AMMA. The analysis will build on and expand some of the concepts in- troduced in this section and provide a consistent framework for the detailed analysis of DACCIWA field campaign data in following years. From an atmospheric dynamics and chem- istry perspective, SWA is of particular interest and has not been studied much in the past. AMMA had fewer stations in SWA and only a few publications covering this region.

During GATE, data quality (e.g. from radiosondes and satel- lites) and data assimilation (e.g. use of cloud motion vec- tors) had not evolved enough to allow a reliable analysis of 850 hPa streamlines for example, especially over the Gulf of Guinea (e.g. Sadler and Oda, 1979). Relying on the dens- est radiosonde network at the Guinea coast and in the Suda- nian zone (Côte d’Ivoire, Ghana, Togo, Benin, Nigeria) ever, DACCIWA can for the first time provide a detailed account of SWA weather systems and their impacts on precipitation and atmospheric composition. In contrast to the Sahel, SWA is often characterised by situations with high moisture and relatively low convective inhibition (CIN), while the vertical wind shear is typically weak. Thus, convection is relatively easy to trigger and remains less well organised, yet brings substantial rains. These are often connected to weak and ver- tically shallow cyclonic and anticyclonic vortices (e.g. Fink et al., 2006), but details of this relationship are still unclear as is their linkage to classical equatorial wave and AEW distur- bances. This paper will shed some new light on these funda- mental unexplored dynamical features, including the specific question of onset dynamics.

The paper is structured as follows: in Sect. 2 an overview

of the employed data and methods will be given. Section 3

contains a relatively short discussion of the large-scale set-

tings followed by a more detailed analysis of the synoptic-

scale evolution in Sect. 4. Section 5 discusses the implica-

tion of meteorological variation on atmospheric composi-

tion, focusing on Saharan and Sahelian dust, biomass burn-

ing aerosol from the SH and pollution plumes from the cities

along the Guinea coast. Main conclusions will then be given

in Sect. 6.

C

ON

L

Niger

Benue Niger

Lake Volta Lac de

Kossou

A T L A N

T I C O

C E A N

B^IGHTOF BENIN

G

G

GO L D C^{O A}

S T Cotonou

AccraLomé Abidjan

Kumasi Lamto

Parakou

Monrovia Conakry

Freetown

Ouagadougou

Abuja

Yaoundé Malabo

PE PP E R C^{O A S} T

Guinea Benin

Cameroon

Ile Ife Save

O_SHOGBO H_ILLS F^OUTA

D^JALON

J_O S P_LATEAU

G

A

H

IGHLA NDS

C^HAÎNE DE L'A^TAK

ORA

Figure 1. Geographical overview of the study region with sites and names. The purple rectangle marks the main DACCIWA focus region (5–10

N, 8

W–8

E).

2 Data and methods 2.1 Data

For the investigation of atmospheric dynamics, analysis and reanalysis data from the European Centre for Medium-Range Weather Forecasts (ECMWF) are used. Most of the analy- ses are based on the ERA-Interim (hereafter ERA-I) reanal- ysis at about 0.7

grid spacing (Dee et al., 2011), which al- lows the computation of background climatologies back to 1979. For investigations focusing on the campaign period in 2016 alone, the higher-resolution operational analyses (na- tive resolution of ∼ 9 km; model version Cy41r2; see www.

ecmwf.int) are employed. As there was no change to the op- erational system during the study period, these data can be regarded as homogeneous, in contrast to longer time spans of operational data. The majority of radiosondes launched dur- ing the DACCIWA field campaign were distributed through the Global Telecommunication System and were assimilated at ECMWF. For the analysis of ocean influences on West Africa, the 0.25

daily Reynolds Optimum Interpolation SST data are used. The dataset combines observations from dif- ferent platforms (satellites, ships, buoys) on a regular global grid. A spatially complete SST map is produced by interpo- lating to fill in gaps (Reynolds et al., 2007). Data have been retrieved from the NOAA NCDC (National Oceanic and At- mospheric Administration – National Climatic Data Center) FTP site (http://www.ncdc.noaa.gov). Monthly anomalies for June–July 2016 and daily anomalies are based on the 1981–

2016 climatology.

As a precipitation estimate, the standard Tropical Rain- fall Measuring Mission (TRMM) product 3B42 (v7) with 0.25

grid spacing is used (Huffman et al., 2007). This prod- uct combines information from space-borne radar and mi- crowave and infrared channels, subject to monthly calibra- tion with surface rain gauges if available. Since September 2014, the real-time calibration of microwave radiances using

the precipitation radar has ceased due to the decommission- ing of the TRMM satellite and was replaced by using cli- matological adjustments. Although this caused a discontinu- ity, the TRMM 3B42 product was prioritised over the Global Precipitation Measurement (GPM) Integrated Multi-satellite Retrievals for GPM (IMERG) successor product due to the longer availability (1998–2016), which allowed for the cal- culation of anomalies. The temporal resolution of this prod- uct is once every 3 h, but here daily accumulations (22:30–

22:30 UTC) are used for most investigations. In addition, out- going longwave radiation (OLR) data from the Spinning En- hanced Visible and Infrared Imager (SEVIRI) on the geosta- tionary Meteosat Second Generation (MSG) satellites with a spatial and temporal resolution of ∼ 3 km at nadir and 15 min, respectively, are used as a proxy for convective ac- tivity (Schmetz et al., 2002). In particular, channel 9 (i.e. ap- proximately 9.80–11.80 µm) of the thermal infrared band is taken to retrieve cloud-top temperatures and to ensure day and night coverage. Different types of clouds are analysed using information on cloud-top characteristics (CTX) and the cloud mask (CMA) from the Satellite Application Facility on Climate Monitoring (CM SAF). Both the CTX and CMA subsets are part of the CLAAS-2 (Finkensieper et al., 2016) dataset, which is derived from information provided by SE- VIRI (Stengel et al., 2013). Therefore, CLAAS data have the same temporal and spatial resolution as the SEVIRI dataset.

As this paper is meant to give a broad overview of me-

teorological and chemical conditions only, a detailed anal-

ysis of DACCIWA field campaign data is left to follow-up

studies. The only exception is radiosonde data from Abid-

jan (for location, see Fig. 1) used to illustrate a period of

unusual dryness during July 2016. Relative humidity was de-

rived from (00:00, 06:00, 12:00 and 18:00 UTC) soundings

four times daily using the high-resolution vertical profiles

obtained from the MODEM radiosonde system. The analysis

will concentrate on the main DACCIWA study region (8

W–

8

E, 5–10

N, see Fig. 1), but influences on that region from a much wider area will be considered.

2.2 Methods

In order to better understand and characterise atmospheric variability during the DACCIWA campaign, a number of fea- tures important for SWA were objectively or subjectively identified:

1. Equatorial waves: the presence of CCEWs is identi- fied using the wave filtering method in specific wave- number–frequency domains as described in Wheeler and Kiladis (1999). In addition to the CCEWs (Kelvin waves, MJO, mixed Rossby-gravity and equatorial Rossby waves), tropical depression-like disturbances (TD) are filtered following the method by Roundy and Frank (2004). These often correspond to AEWs over West Africa. The filtering is applied to the 3-hourly TRMM 3B42 (v7) precipitation dataset (see Sect. 2.1) within the northern equatorial band 5–15

N, which contains the bulk of the convective precipitation during the campaign period but excludes some heavier oceanic rainfalls (see Fig. 5b).

2. Heat low index: following Lavaysse et al. (2009), the low-level atmospheric thickness between 925 and 700 hPa over a domain that covers northern and West Africa is used to determine the location and the inten- sity of the SHL. The location corresponds to the region with thickness values larger than the 90th percentile.

The intensity is defined directly through the thickness in geopotential metres (gpm), indicating the thermal dila- tion of the lower atmosphere. Once the SHL is detected and the intensity of each grid point calculated, the cen- tre of the SHL is defined as the barycentre in longitude and latitude, which is closely linked to the east and west phases of temperature anomalies proposed by Chauvin et al. (2010) when the SHL is located in its Saharan location (from end of June to mid-September). These computations are based on ERA-I (see Sect. 2.1).

3. AEJ index: average position and strength of the AEJ are objectively calculated based on Berry et al. (2007).

Within the region 0–30

N and 8

W–8

E (longitudinal extent of DACCIWA focus region; see Fig. 1), 6-hourly ERA-I winds at 700 hPa are used. A spatial low-pass filter with a cut-off wavelength of 1000 km is applied to calculate shear vorticity, which is then used to determine the jet axis. The average wind speed along the jet axis and the mean latitudinal position is estimated for June–

July 2016 and the long-term climatology (1987–2016) for comparison.

4. Mesoscale convective system (MCS) identification: the evolution of deep convective clouds is monitored by applying an overlap-based tracking algorithm (Mathon

and Laurent, 2001; Schröder et al., 2009; Williams and Houze, 1987) to the 15 min infrared data of SEVIRI.

In two successive images, cold cloud regions are iden- tified first and then connected in time by determining the highest accordance with respect to area, area over- lap and spatial translation. Here, deep convective clouds are defined as regions with a brightness temperature of

≤ 233 K and an area of least 100 contiguous pixels (i.e.

∼ 900 km

). The former criterion is widely used as a proxy for deep precipitating convection in tropical re- gions, whereas the latter excludes convective systems with low contribution to total cold cloud cover (Mathon and Laurent, 2001; Schröder et al., 2009).

5. Synoptic-scale vortices: close inspection of daily weather charts suggests that only a few classical AEWs occurred during the study period and that a more flex- ible approach is needed to fully represent the observed richness of coherent features. After some testing, a com- bination of subjective tracking of vortex centres from unfiltered 850 hPa streamlines with Hovmöller plots of 850 hPa vorticity and meridional wind was selected.

6. Long-range transport of biomass burning and dust en- riched air masses: biomass burning plumes from central Africa transported into the domain were tracked with carbon monoxide (CO) calculated from the ECMWF Copernicus Atmosphere Monitoring Service-Integrated Forecasting System (CAMS-IFS; Inness et al., 2013).

Dust plumes from Sahelian and Saharan sources, north of the DACCIWA domain, were identified using the CAMS-IFS dust aerosol optical depth (DAOD). Where available, CAMS-IFS assimilates satellite information to bring the model output closer to reality.

7. Turbulent dispersion of urban plumes from the five ma-

jor cities, where DACCIWA aircraft and ground mea-

surements were taken (Abidjan, Kumasi, Accra, Lomé,

Cotonou; see Fig. 1), were calculated daily using for-

ward trajectories of passive tracers in a Lagrangian

framework. Two models were used: (a) FLEXPART

v6.2 (https://www.flexpart.eu/; Stohl et al., 2005) driven

with ECMWF ERA-I winds and (b) HYSPLIT v4.8

(Stein et al., 2015) driven by GDAS (Global Data As-

similation System) winds. Both models were run for

24 h with continuous emissions of the tracer. The extent

of the plume was calculated in FLEXPART using the

root mean square distance of particles from the source at

the end of the 24 h simulation. For HYSPLIT the plume

boundary was defined at the end of the 24 h at a thresh-

old concentration of 10

m

, with one unit of tracer

being emitted from the source in 24 h.

°C

70° N 60° N

40° N 30° N 20° N 10° N 50° N

10° N 20° N

40° N 50° N 60° N 70° N 30° N EQ

180° 120° W 60° W 0° 60° E 120° E 180°

Figure 2. Global SST anomalies for June–July 2016 (

C). Basis is the Reynolds Optimal Interpolated SST v2 dataset and anomalies are relative to 1981–2016. Only anomalies above the 95 % confi- dence level based on a two-sided Student’s t test are plotted. The black box marks the area used for Fig. 3.

3 Large-scale settings

This short section aims to characterise the large-scale set- ting the DACCIWA field campaign period, June–July 2016, was embedded in. Figure 2 shows global SST anomalies for June–July 2016. While at the beginning of the year El Niño conditions were still prevalent (not shown), by June a tran- sition to La Niña had occurred, which usually favours mon- soonal precipitation in the Sahel (Joly and Voldoire, 2009).

At the same time, the equatorial Atlantic Ocean was rela- tively warm with widespread anomalies above 1 K (Fig. 2).

These warm events, sometimes referred to as Atlantic El Niños (Okumura and Xie, 2006), are associated with a sup- pressed ACT and are linked with westerly surface wind per- turbations at the Equator. The reduced surface wind stress causes less surface oceanic divergence and vertical mixing, leading to reduced SST cooling. This reduces the pressure gradient towards the SHL and thus the inland penetration of monsoonal rains. Since the 1970s, a frequent anticorre- lation between El Niño in the Atlantic Ocean and El Niño in the Pacific Ocean has been observed (Rodríguez-Fonseca et al., 2015).

Warmer equatorial waters in the Gulf of Guinea as in 2016 exhibit a strong correlation with above-normal rainfall at the Guinea coast, which has been robust throughout the 20th cen- tury (Diatta and Fink, 2014). Mohino et al. (2011) argue that a warm eastern equatorial Atlantic Ocean and a simultaneous cold eastern Pacific Ocean exert compensational forcings on Sahelian rainfall, such that the archetypical dipole response during warm years in the Gulf of Guinea has rarely been ob- served after the 1970s. In the Mediterranean Sea, positive SST anomalies are found over the eastern basin accompa- nied by negative anomalies in the northwestern part of the In- dian Ocean (Fig. 2). Positive SST differences between these two areas are associated with rainfall excess over the Sahel (Fontaine et al., 2011; Park et al., 2016). Overall, it appears that a combination of these different factors was in place in

6° N

4° N

2° N

2° N

4° N

6° N

8° N

10° N EQ

1 Jun 2016

6 Jun 11 Jun 16 Jun 21 Jun 26 Jun 1 Jul 6 Jul 11 Jul 16 Jul 21 Jul 26 Jul

Figure 3. Daily SST behaviour over the eastern tropical Atlantic during June–July 2016. SSTs (

C) are averaged between 10

W and 4

E (see box in Fig. 2) and shown as absolute values (lines) and anomalies (shading). Basis is the Reynolds Optimal Interpo- lated SST dataset and anomalies are relative to 1981–2016 as in Fig. 2. Only values greater than the SST Reynolds daily error esti- mation are plotted. The four phases of the DACCIWA campaign are marked with thin green lines.

2016 since the June–September Sahelian rainfall was only very slightly above normal (not shown). Standard indices in- dicate that the MJO was not active over West Africa in June–

July 2016 (not shown).

For the DACCIWA focus region (Fig. 1) and along most of the Guinea coast, June–September rainfall turned out to be normal (not shown), despite the extended dry spell during Phase 2 discussed in Sect. 4 (Fig. S1b in the Supplement).

Only the Guinea and Cameroon line mountains and the Bight of Bonny had above-normal rainfall. It is unclear why the warmer waters in the Gulf of Guinea did not cause more rain- fall in lowland areas. A possible explanation is warm SSTs in the South Atlantic (Fig. 2), which seem to reduce the positive effect of the equatorial Atlantic (Nnamchi and Li, 2011). In 2016, the situation was further complicated by relatively cold SSTs along the coasts of Senegal, Ghana and Togo, whose impacts on rainfall are not clear.

An inspection of the daily evolution of zonally averaged

(10

W–4

E) SSTs over the tropical eastern Atlantic during

June–July 2016 reveals the typical establishment of the equa-

torial cold tongue (1

N–5

S) and of upwelling of cooler wa-

ter along the Guinea coast (4–6

N) (Fig. 3). The onset of

the ACT occurs around 10 June 2016 (mean date is 11 June

with a standard deviation of 12 days according to Caniaux

et al., 2011) with SSTs slightly below the long-term average,

followed by a significant warming and southward retreat be-

tween mid-June and 5 July 2016, with warm anomalies sur-

passing 1.5 K. After that, the ACT is re-established but abso-

lute temperatures stay above average by 0.5 K or more until the end of July 2016, consistent with the anomalies shown in Fig. 2. The coastal upwelling sets in much later (Fig. 3). First indications of a cooling are found around 18 June 2016, but a more substantial cooling begins on 27 June 2016 until SSTs drop below 26

C across a broader coastal strip until the end of July. The phenomenon is a little stronger in 2016 than in other years, with negative anomalies of the order of 0.5 K, particularly along 5

N.

The other important driver of the WAM is the SHL. Fig- ure 4 shows its intensity and position on a daily basis during June–July 2016. During the first 3 weeks of June, the SHL is in an intense phase (Fig. 4a) and shows large east–west fluc- tuations with a period of about 10 days, remaining mostly to the east of the climatological position (Fig. 4b). This shows some resemblance with the SHL variations described by Chauvin et al. (2010) but on rather short timescales. The SHL is also located further to the north than usual, associated with large positive temperature anomalies over northeastern Africa and anomalous southwesterly flow over the eastern Sahel (not shown). Around 20 June, the SHL abruptly weak- ens and shifts to a more southerly position, followed by a gradual intensification and northward retreat during the fol- lowing week (Fig. 4a). It is still located east of its clima- tological position during this period (Fig. 4b). After that, a long strong phase begins to last until 18 July (Fig. 4a), during which the SHL gradually shifts westward (Fig. 4b) and also slightly northward. Around 18 July another abrupt weaken- ing occurs and continues until the end of July 2016, only shortly interrupted by positive values (Fig. 4a). The SHL is located to the west of its climatological position during this time (Fig. 4b) and then migrates to the south at the end of July (Fig. 4a). In the next section, the impact of these fluc- tuations and those of SSTs on synoptic-scale variability over SWA will be discussed.

4 Detailed synoptic analysis 4.1 General approach

In order to guide the discussion of the DACCIWA field mea- surements, the study period is divided into distinctive phases and the most significant weather systems are labelled for better reference in other papers. The division into phases is mainly based on the north–south precipitation differ- ence (NSPD hereafter) between the coastal zone (0–7.5

N) and the Sudanian–Sahelian zone (7.5–15

N), both averaged across the longitude range 8

W–8

E (see Fig. 6 for orien- tation). Figure 5a shows daily values of the NSPD based on TRMM precipitation estimates for June–July 2016. Figure 5b shows the corresponding zonally averaged rainfall values against latitude. Four distinct phases are recognisable from this analysis.

(a) (b)

39° N 36° N 33° N 30° N 27° N 24° N 21° N 18° N 15° N 12° N

1 Jun 6 Jun 11 Jun 16 Jun 21 Jun 26 Jun 1 Jul 6 Jul 11 Jul 16 Jul 21 Jul 26 Jul

1 Jun 11 Jun 21 Jun

Date Longitude

1 Jul 11 Jul 21 Jul 31 Jul

Figure 4. SHL evolution during June–July 2016. (a) Time–latitude anomalies of the SHL intensity (gpm) defined as the thickness be- tween 925 and 700 hPa (relative to 1979–2016). (b) Longitudinal location of the SHL barycentre (black line) with the 1979–2016 percentiles in colour shading. See Sect. 2.2 for more details. The four phases of the DACCIWA campaign are marked with thin black lines.

Phase 1 lasts from 1 to 21 June 2016 and is characterised by a rainfall maximum near the coast. However, it shows large fluctuations with periods around 5 days (Fig. 5a). Par- ticularly the middle part of Phase 1 is very wet, while the earlier and later parts are characterised by more isolated rain- fall peaks near 4

N (Fig. 5b). This modulation is consis- tent with the QBZD index, showing a significant minimum around 14 2016 (see http://misva.sedoo.fr). Rainfall during this period is unusually intense offshore of the Niger Delta area stretching across the Gulf of Guinea towards Cape Pal- mas (Fig. 6a; see also Fig. S1a for anomalies). A second rain- fall maximum is located over the tropical Atlantic to the west of West Africa in Fig. 6a, where SSTs are climatologically much warmer (not shown). Precipitation does already stretch far inland into the Sahel but amounts are relatively low with the exception of the Cameroon line highland region along the border of Nigeria and Cameroon. The moderate changes from drier to wetter and back to drier conditions in the Sahel during Phase 1 are reflected in weak but hardly significant undulations of the intra-seasonal Sahelian index reaching a minimum on 12 June 2016 (see http://misva.sedoo.fr). The pre-monsoonal conditions are also reflected in fields of zon- ally averaged total column water vapour (TCWV) with val- ues above 45 mm mostly restricted to south of 12

N (Fig. 7).

The ITD (identified by the 14

C isoline of 2 m dew point) fluctuates around 16

N (Fig. 7).

Phase 1 corresponds closely to the period of anomalously

strong SHL that fluctuates from east to west discussed in

Sect. 3 (see Fig. 4), while correspondence to SST behaviour

in the Gulf of Guinea (Fig. 3) is less clear. It is interest-

Figure 5. Rainfall evolution during June–July 2016. (a) North–south precipitation difference based on the 7.5–15 and 0–7.5

N bands (see boxes in Fig. 6). (b) Latitudinal distribution of rainfall. Both panels are based on daily TRMM precipitation values averaged over 8

W–

8

E (longitudes bordering DACCIWA focus region; see Fig. 1). The four phases of the DACCIWA campaign and significant synoptic-scale features A–J are marked at the approximate time (and also latitude in b) of crossing the DACCIWA focus region.

Figure 6. Horizontal distribution of mean precipitation during the four phases of the DACCIWA campaign. Plots are based on TRMM precipitation and given in millimetres per hour. (a) Phase 1 (1–21 June 2016), (b) Phase 2 (22 June–20 July 2016), (c) Phase 3 (21–26 July 2016) and (d) Phase 4 (27–31 July 2016). The black boxes mark the areas used to compute the north–south precipitation difference shown in Fig. 5a. Corresponding anomalies are shown in Fig. S1.

ing to note that despite a strong SHL and an established ACT, rainfall remains strongest along the coast, indicating that monsoon onset has not yet occurred. This aspect will be discussed further in Sect. 4.3. In Fig. 5, significant synoptic disturbances are labelled with the capital letters A–J. These

were subjectively identified from 850 hPa streamline plots at

00:00 UTC each day and often show a noticeable correspon-

dence to the precipitation and TCWV behaviour. A summary

of their most important characteristics is given in Table 1 and

individual tracks are shown in Figs. 11, 13, 14 and 16. The

Figure 7. Evolution of TCWV (shading in millimetres) and the ITD (black line, identified from the 14

C isoline of 2 m dew point) during June–July 2016 based on ECMWF operational analysis. The four phases of the DACCIWA campaign and significant synoptic-scale features A–J are marked at the approximate time and latitude of crossing the DACCIWA focus region as in Fig. 5b.

locations of the feature labels in Figs. 5 and 7 correspond to the times when they cross the Greenwich meridian (i.e.

centre of the DACCIWA focus region) and their latitudinal position (Figs. 5b and 7 only).

Phase 2 lasts from 22 June to 20 July 2016 and is charac- terised by a rainfall maximum inland with smaller and less regular fluctuations of the NSPD (Fig. 5a) and only occa- sional and weaker convective systems around 4

N (Fig. 5b).

This indicates a fully developed WAM with a deeper pen- etration of rainfalls and TCWV into the continent, and a northward-shifted ITD, while marine precipitation is re- stricted to the Bight of Bonny and the waters along the West African west coast (Figs. 6b and 7). Large parts of the in- land DACCIWA region were virtually dry during this period, much drier than in other years (Fig. S1b), despite relatively high TCWV values (Fig. 7). The transition from Phase 1 to Phase 2 is marked by strikingly dry conditions across most of the area of interest (Fig. 5b), much reduced TCWV (Fig. 7), strong fluctuations of the ITD (Fig. 7) and an abrupt break- down of the SHL (Fig. 4a). During Phase 2 the SHL then gradually intensifies and shifts westward (Fig. 4). There is also a gradual increase in coastal upwelling during this pe- riod (Fig. 3), which is consistent with more stable, near- surface monsoonal winds. The behaviour of the ACT, which is relatively weak and shifted to the south during most of Phase 2, does not seem to be closely related to the precipita- tion shift.

During 21–26 July 2016 (Phase 3), the rainfall maxi- mum shifts back to the coastal zone (Fig. 5a), accompanied by wet conditions spanning large parts of the latitude band from 1 to 22

N (Fig. 5b), where TCWV is enhanced and the ITD reaches its northernmost extension (Fig. 7). A hor- izontal distribution of rainfall during this period (Fig. 6c) shows unusually intense convection across the entire Gulf of Guinea, widespread rain across the entire Sudanian zone and more patchy local maxima stretching into the Sahel and even southern Sahara (see anomalies in Fig. S1c). Even larger amounts are found along the coast of Guinea and Sierra

(a)

(b)

Phase 2

Phase 3

Phase 4

Figure 8. AEJ evolution during June–July 2016. Time series of (a) latitudinal position and (b) mean speed in metres per second of the AEJ objectively identified from ERA-I reanalysis data. The red lines indicate the 2016 evolution, the blue lines the 1987–2016 climato- logical mean (see Sect. 2.2 for more details). The four phases of the DACCIWA campaign are indicated by vertical lines. The significant synoptic-scale features A–J are marked at the approximate time and latitude of crossing the DACCIWA focus region.

Leone. This wet phase is preceded and accompanied by a second breakdown of the SHL as well as a marked westward shift in its centre (Fig. 4). Coastal upwelling is increased dur- ing this period, while no major change in the ACT is seen (Fig. 3).

During the last 5 days of July 2016 (Phase 4), the WAM

system returns to a more typical behaviour for this time of

the year with a precipitation maximum in the Sahel, similar

to Phase 2 (Fig. 5). As in Phase 2, the southern parts of the

DACCIWA region are rather dry and coastal rainfalls are re-

Table 1. Characterisation of the labelled (Column 1) synoptic-scale features. Columns 2 and 3 give the time period and longitude range for which a coherent vortex was tracked in 850 hPa streamlines (see Figs. 11, 13, 14 and 16). Column 4 gives a general description, including aspects such as propagation speed, latitude and reflection in wind and vorticity anomalies.

Label Time Long Description

A 9–12 June 22

E–25

W fast-propagating (16.8 m s

) Sudanian ( ∼ 11

N) cyclonic vortex and tropical disturbance, weak vorticity signal but clear southerly wind signal, long-lived MCSs embedded B 12–15 June 9

E–21

W moderately fast (11.2 m s

) coastal ( ∼ 5

N) cyclonic vortex and tropical disturbance,

moderate vorticity and meridional wind signals, intense, long-lived MCSs embedded C 15–18 June 11

E–8

W moderately fast (11.0 m s

) Sudanian ( ∼ 9

N) cyclonic vortex and tropical disturbance,

patchy vorticity and meridional wind signals, long-lived MCS embedded

D 15–25 June 17

E–23

W low-pressure trough turning into westward-propagating disturbance with two cyclonic centres, large meridional wind anomalies, relative dryness, triggering monsoon onset E 27–30 June 2

E–27

W AEW with two cyclonic centres and typical propagation speed of 9.6 m s

, with coherent

signals in meridional wind and vorticity, leading to increased Sahelian rainfall

F 29 June–3 July 14

E–21

W AEW with two cyclonic centres and a fast propagation speed of 10.6 m s

, with coherent signals in meridional wind and vorticity, leading to increased Sahelian rainfall

G 3–8 July 12

E–18

W unorganised AEW with ill-defined southern centre, northward moving northern centre and varying propagation speed, but discernable rainfall signal

H 9–16 July 12

E–22

W slowly moving (7.1 m s

) northern ( ∼ 15

N) cyclonic and southern anticyclonic ( ∼ 4

N) vortex originating from SH, westerly wind anomaly between centres I 17–27 July 23

E–25

W relatively slow-moving northern ( ∼ 13

N) cyclonic and southern ( ∼ 5

N) anticyclonic

vortex with westerly wind anomaly in-between, creating conditions for wet period J 23–30 July 19

E–25

W mostly slow moving Sudanian (∼ 9

N) cyclonic vortex, coherent vorticity but less clear

wind signal, occurring in an environment of MCSs, high moisture and widespread rain

stricted to the Niger Delta region (Fig. 6d). Rainfall along the coast of Guinea, however, is even more abnormal than in Phase 3 (Fig. S1d). Overall, conditions are somewhat wetter than during Phase 2. This is accompanied by a partial recov- ery of the SHL (Fig. 4) and weakening of coastal upwelling (Fig. 3).

In the remainder of this section, the four phases outlined above as well as the transition between Phases 1 and 2 (the monsoon onset) will be analysed in detail, focusing on the synoptic-scale features labelled in Fig. 5. To aid the charac- terisation of these features, the following additional diagrams will be considered (see Sect. 2.2 for more details): (a) objec- tive analyses of AEJ position and speed (Fig. 8), (b) Hov- möller plots of 850 hPa vorticity and meridional wind for the 4–18

N latitude band (Fig. 9), and (c) Hovmöller plots of equatorial wave disturbances in the 0–15

N band based on TRMM rainfall as well as tracks of long-lived MCSs (Fig. 10).

4.2 Phase 1: pre-onset (1–21 June 2016)

As stated in Sect. 4.1, the pre-onset period is characterised by a coastal rainfall maximum (Fig. 5), a strong eastward- shifted SHL (Fig. 4) and a weak ACT (Fig. 3). The AEJ is still located close to the coast during most of this phase

(Fig. 8a; see also Fig. S2a). The first week (1–6 June) is relatively quiet with overall little rainfall across the region (Fig. 5b). The AEJ is anomalously far south (Fig. 8a) with a below-normal intensity (Fig. 8b). No significant coherent features are detected during this period, neither in 850 hPa vorticity and meridional winds (Fig. 9) nor in terms of fil- tered equatorial waves (Fig. 10). The enhanced vorticity fea- ture starting on 5 June 2016 (Fig. 9) is related to a northern area of high horizontal wind shear (not shown) and is thus not associated with coherent meridional wind signals. The activ- ity of long-lived MCSs is also relatively weak (black lines in Fig. 10; see also Fig. S3).

Between 7 and 15 June 2016 the AEJ begins shifting

northward, showing two distinct mean speed maxima of

more than 14 m s

(Fig. 8). On 7 June the jet maximum

is located over southern Chad (not shown). The enhanced

shear associated with this feature appears to have supported

the formation of a large and long-lived MCS (Fig. 10) that

brings substantial rainfall to southern areas (Fig. 5b) and

thus creates a minimum in NSPD (Fig. 5a). In the follow-

ing days, three relatively weak cyclonic disturbances cross

the region (Fig. 11). As already mentioned, Table 1 provides

a summary of the main characteristics of these disturbances

and all subsequent ones. The first disturbance (labelled A

30° W 20° W 10° W 0° 10° E 20° E

Ju n Ju l

2.0 1.5 1.0 0.5 0.0 -0.5 -1.0 -1.5 -2.0

Figure 9. Coherent wind and vorticity features affecting the DACCIWA region. Hovmöller diagram showing 4–18

N meridionally averaged vorticity (colours, 10

s

) and meridional wind (black lines, m s

) based on operational ECMWF analyses at 1

resolution (in order to smooth noisy vorticity fields). The four phases of the DACCIWA campaign and significant synoptic-scale features A–J are marked as well as the longitudinal bounds of the DACCIWA focus region 8

W–8

E (see Fig. 1).

in Fig. 11) propagates quickly westward from eastern Chad to northern Côte d’Ivoire between 9 and 11 June 2016, in accordance with the relatively strong AEJ during this pe- riod (Fig. 8b). When it passes the DACCIWA region, the in- crease in southerly flow seen in Fig. 9 (solid black lines) is associated with an increase in rainfall inland, while coastal rainfall is also still active, leading to an NSPD near zero (Fig. 5a). The vorticity signature of Feature A is relatively weak (Fig. 9), but there are several long-lived MCSs embed-

ded in this system and the latter stages are identified as a TD (Fig. 10).

The immediately following second disturbance (labelled B

in Fig. 11) propagates a little slower and on a more southern

track from the border of Nigeria and Cameroon parallel to

the coast out to the Atlantic past 20

W. When the centre of

the vortex passes the DACCIWA region on 12 and 13 June, a

strong increase in rainfall over the ocean is observed, creating

a sharp minimum in NSPD (Fig. 5a). The slower propagation

of this feature is consistent with the larger distance to the

30° W 20° W 10° W 0° 10° E 20° E

Figure 10. Tropical wave phenomena and long-lived MCSs during June–July 2016. Hovmöller diagram of 5–15

N meridionally averaged precipitation from TRMM (mm h

, colour shading according to legend) with objectively identified waves marked with coloured lines according to the legend in the top right corner and long-lived MCSs with at least 24 h of lifetime marked with thick black lines (for details on detection of both features, see Sect. 2.2). Contour lines for the wave features correspond to a modulation of precipitation of more than 0.12 mm h

. Note that while the tropical waves are identified for the entire longitudinal range of 35

W–25

E, the MCS identification is limited to the land-dominated area 20

W–25

E. The four phases of the DACCIWA campaign and significant synoptic-scale features A–J as well as the longitudinal bounds of the DACCIWA focus region 8

W–8

E (see Fig. 1) are marked.

strong AEJ core near 9

N (Fig. 8). Feature B shows a more coherent signature in vorticity and meridional wind (Fig. 9), as well as TCWV (Fig. 7), and is identified as a TD with

two very long-lived and intense MCSs embedded over the DACCIWA region (Fig. 10).

From 15 to 18 June 2016 the AEJ is weak and shifts north-

ward (Fig. 8), while a third cyclonic feature becomes ev-

10 09 12 11

15 14

13

12 16 15 18 17

18° N

10° N

4° N

Figure 11. Significant synoptic-scale features during 9–18 June 2016 (Phase 1). To create this graph, vortices were subjectively identified in 850 hPa streamlines based on operational ECMWF analyses. All vortex positions refer to 00:00 UTC with the dates given as numbers. Round symbols mark cyclonic systems (labelled A, B and C). Paler colours are used for days when the vortices were not clearly identifiable. The boxes mark the areas used to compute the NSPD shown in Fig. 5a. The stippled lines show the latitude range used to produce Fig. 9.

ident in the 850 hPa streamlines (labelled C in Fig. 11). It propagates relatively slowly from eastern Nigeria across the DACCIWA region, reaching northwestern Côte d’Ivoire by 00:00 UTC on 18 June. It is associated with a moderate in- crease in rainfall inland, while the coast is conspicuously dry, leading to a slightly positive NSPD (Fig. 5a). Interestingly, Feature C appears to be related to a longer-lived, somewhat patchy vorticity and meridional wind feature (white line in Fig. 9), which is also identified as a TD (Fig. 10). The vortex identified from the streamlines appears to move somewhat slower than this disturbance and is only found during the period of strongest southerlies immediately over the DAC- CIWA region (not shown). The vorticity feature moves at a similar speed as the embedded MCS (Fig. 10). Interest- ingly, during the middle part of Phase 1, the rainfall in the 5–

15

N latitude band appears to be modulated by two Kelvin waves propagating across the DACCIWA region (green lines in Fig. 10), which superpose with the TD signals. There is also some indication for equatorial Rossby wave activity in the western part of the domain, but this signal is harder to see in the unfiltered TRMM data (Fig. 10).

4.3 Transition from Phase 1 to Phase 2: the onset (16–26 June 2016)

The monsoon onset is often defined as a more permanent shift in the rainfall maximum into the continent (e.g. Fitz- patrick et al., 2015). According to the NSPD (Fig. 5a), this occurred on 21–22 June in 2016, the transition from Phase 1 to Phase 2. As this date is of such a large importance for the WAM, a dedicated discussion of the 5 days before and af- ter this date is presented here. Overall this 10-day period has relatively low rainfall in the DACCIWA region, the two note- worthy exceptions being the enhanced coastal rainfall around 19 and 20 June and a Sahelian maximum on 24–25 June

40° W

20° W

°

60° W 30° W 0° 30° E

Figure 12. Extratropical influences during the monsoon onset.

Shown are streamlines coloured by wind speed (scale at bottom), at 600 hPa and mean sea level pressure (grey shading) at 00:00 UTC on 17 July based on ECMWF operational analyses. The two distur- bances from Fig. 13 are marked in red.

18 15

24 25

22

19 16

17

19

21 20 22

23

24 23

21

20 19

18 17

20 17

18° N

10° N

4° N

Figure 13. Significant synoptic-scale features during 15–25 July 2016 (transition from Phase 1 to Phase 2). To create this graph, vortices were subjectively identified in 850 hPa streamlines based on operational ECMWF analyses. All vortex positions refer to 00:00 UTC with the dates given as numbers. Round symbols mark cyclonic vortices (labelled D

, D

and D

) and the thick orange line shows the southernmost extension of a significant trough at 600 hPa (see Fig. 12). Vortices with a joint propagation are linked with dashed grey lines. The boxes mark the areas used to compute the NSPD shown in Fig. 5a. The stippled lines show the latitude range used to produce Fig. 9.

(Fig. 5b). Consistently, the activity of equatorial waves and long-lived MCSs is strongly suppressed (Fig. 10).

The synoptic development during the onset is charac-

terised by substantial extratropical influences disturbing the

circulation over northern Africa with high-amplitude waves

and wave breaking in the subtropical upper troposphere. On

15 June 2016, the polar and subtropical jets merge over the

Mediterranean Sea and a high-amplitude ridge is located up-

stream over the central North Atlantic (not shown). This pe-

riod is characterised by a substantial drop in the inertial sta- bility index defined by Cook (2015) but negative values are only reached for short periods (not shown). While the At- lantic ridge continues to propagate eastward, the downstream trough and ridge amplify strongly until 17 June. On this day, the trough stretches all the way to the Mauritanian coast and leads to a strong southwesterly flow at 600 hPa across the western Sahara (Fig. 12). It is conceivable that subsidence associated with the ridge stretching from eastern Europe into northeastern Africa (Fig. 12) contributed to the suppression of rainfall evident from Fig. 5b. The inflow of cool maritime air from the Atlantic Ocean leads to an abrupt ventilation of the SHL, causing a weakening of its intensity and rapid east- ward shift in the centre (Fig. 4) with some resemblance to the situation Todd et al. (2013) refer to as the “maritime phase”.

According to http://misva.sedoo.fr, the intra-seasonal SHL index reached a distinct maximum on 17 and 18 June 2016.

The extreme nature of this cold surge is visible in 20

W–0

E averaged temperature anomalies at 850 hPa, showing a very distinct and unusual cooling during this period, with anoma- lies below − 6 K north of 30

N and below − 4 K down to almost 20

N (Fig. S4a). Over the next few days, the whole wave slowly drifts eastwards, allowing the northerlies asso- ciated with the trough to penetrate into the eastern parts of the Sahara as well. The associated cooling, which is overall less dramatic in the east (Fig. S4b), finally leads to the con- spicuous collapse of the SHL between 20 and 25 June 2016 shown in Fig. 4a. This development is also reflected in an abrupt northward jump of the AEJ core accompanied by a significant weakening around 21 June 2016 (Fig. 8).

Within the large area of reduced surface pressure to the southeast of the trough that stretches unusually far south (Fig. 12), three cyclonic vortices form at 850 hPa (Fig. 13).

The first one (labelled D

) first appears over the Hoggar Mountains in southern Algeria on 15 June 2016 and then re- mains rather stationary over northern Niger between 16 and 18 June 2016. The second (labelled D

) forms farther to the east and slowly moves along the border between Chad and Niger between 17 and 19 June 2016 (see also Fig. 12). The last one (labelled D

) is only discernable in 850 hPa stream- lines on 19 and 20 June 2016 along the border of northern Chad and Sudan. On these two days, the three centres form a zonally elongated area of cyclonic rotation with marked northerlies to the west and southerlies to the east. Between 19 and 21 June 2016, D

and D

rotate cyclonically around each other while beginning to propagate westward in a fash- ion similar to an AEW (Fig. 13). Both cyclonic centres slow down and weaken between 22 and 25 June 2016 near the West African west coast. To the northwest of D

and D

, northerly flow reaches values of 15–25 m s

between 18 and 21 June 2016 (not shown), which leads to a marked south- ward push of TCWV (Fig. 7).

Figure 9 shows how this unusual development is reflected in 850 hPa vorticity and meridional wind (labelled D). On 18 June 2016, D

and D

are still located to the north of 18

N

and thus signals in the Hovmöller plot are weak. On 19 June 2016, the strong northerlies begin to penetrate into the DAC- CIWA region, helping to suppress rainfall (see Fig. 5b). This is followed by unusually large vorticity values on 20 June 2016, when D

moves south of 18

N (Fig. 13). On 20 and into 21 June 2016 a wide area of very strong southerlies spreads across the DACCIWA region. These bring moisture far into the continent, shifting the ITD northward (Fig. 7), and thus create the conditions for an inland rainfall maxi- mum between 23 and 25 June 2016, indicating that the onset has in fact occurred. After the turbulent transition phase, the WAM system becomes relatively quiet and the AEJ slowly gets re-established near its climatological latitudinal position until 26 June 2016 (Fig. 8) with the SHL also beginning to re-intensify (Fig. 4).

The analysis above strongly suggests that in 2016 the mon- soon onset was triggered by very strong interactions with the midlatitudes that supported a suppression of rainfall over West Africa. Low-rainfall conditions around the onset have been documented for other years as well (Sultan and Janicot, 2003).

4.4 Phase 2: post-onset (22 June–20 July 2016)

Phase 2 comprises a period of relatively undisturbed mon- soon conditions. The entire DACCIWA aircraft campaign fell into this period (see Flamant et al., 2017). The NSPD is positive through most of this phase and is modulated by the significant weather systems E–I, with centres between 12 and 16

N and thus farther north than the Phase 1 features A–C (see Fig. 5). This period was anomalously dry along the Guinea Coast (Fig. S1b). Features E–I also modulate the speed and latitude of the AEJ (Fig. 8).

At the very beginning of this phase, between 23 and 26 June 2016, while the monsoon is still being established, a relatively weak (and therefore unlabelled) cyclonic feature crosses the southern part of SWA, creating some moderate rainfalls in the Sahel around 24 June 2016 (Fig. 5b). This is the first time during the DACCIWA campaign that the precip- itation maximum has fully shifted inland. After this system, the SHL starts intensifying and shifts northward (Fig. 4a), the AEJ accelerates (Fig. 8b) and a deep southwesterly monsoon flow gets established (not shown).

Between 27 June and 8 July 2016, three AEWs (Fig. 14a) associated with moderate fluctuations in TCWV (Fig. 7) de- velop. The first two (Features E and F) form on a relatively well organised AEJ (Fig. 8b) and show rather classical prop- agation characteristics with coherent signals in meridionally averaged vorticity and meridional wind at 850 hPa (Fig. 9).

Both have two cyclonic centres at 850 hPa straddling the jet

to the north and south (Fig. 14a) and are also objectively

identified as TDs (Fig. 10). Feature E, which forms near the

Greenwich meridian, appears to be associated with the slight

rainfall enhancement on 26 and 27 June 2016, while Feature

F forming near 12

E creates a peak in rain in the northern

14

15

13 12 11

10 09

16 14

13 12

12 14 13 18° N

10° N

4° N

(b)

30 03

03

29

28 27

05 04 07 08

30 29 01

02 18° N

10° N

4° N

(a)

Figure 14. Significant synoptic-scale features during 27 June–16 July 2016 (beginning (a) and middle (b) of Phase 2). To create these graphs, vortices were subjectively identified in 850 hPa streamlines based on operational ECMWF analyses. All vortex positions refer to 00:00 UTC with the dates given as numbers. Round symbols mark cyclonic systems (labelled E, F, G and H

), squares anticyclonic systems (labelled H

). Vortices with a joint propagation are linked with dashed grey lines. The cores of significant 850 hPa jets are indicated with light blue arrows, again with the date at 00:00 UTC given as numbers. The boxes mark the areas used to compute the NSPD shown in Fig. 5a. The stippled lines show the latitude range used to produce Fig. 9.

box on 29 June 2016 (Fig. 5b). Finally, Feature G forms dur- ing a period when the AEJ weakens and becomes more frag- mented, which makes its latitudinal position vary strongly (Fig. 8). This leads to less clear and slower propagation be- haviour (Fig. 14a). The northern centre propagates from cen- tral Niger to eastern Mauritania between 3 and 6 July 2016 and then drifts northwestward towards the border with West- ern Sahara. This behaviour is accompanied by a rapid shift in the SHL to the west (Fig. 4b). The southern centre shows a less coherent propagation. This is consistent with the rela- tively patchy signals in wind and vorticity shown in Fig. 9, apart maybe from the final stages over the open Atlantic Ocean. Feature G is also not matched with a TD like the previous features are (Fig. 10). Nevertheless, a marked in- crease in rainfall is observed when this feature crosses the DACCIWA region on 4 July 2016 (Fig. 5b).

After that, between 9 and 16 July 2016, a fundamentally different and quite unusual development occurs. While in the north, a cyclonic feature slowly tracks from eastern Mali to Cabo Verde between 8 and 13 July 2016 and then out to the Atlantic (Feature H

in Fig. 14b), there is no clear cor- responding southern vortex. Instead, an anticyclonic system (H

) slowly propagates from Gabon on 11 July across the tropical eastern Atlantic, reaching the coast of Sierra Leone

Jul Jul Jul Jul Jul Jul Jul Jul Jul Jul

Figure 15. Vertical structure of the atmosphere during part of Phase 2. Shown are relative humidity (shading according to scale, four times daily), wind (barbs) and 600–925 hPa vertical wind shear (below main plot) (both two times daily) from radiosondes launched at Abidjan (for location see Fig. 1) during 7–16 July 2016.

on 14 July 2016, after which it begins to weaken over the ocean to the west. As this system moves a little faster than its cyclonic counterpart to the north, the two centres approach each other, creating an area of marked low- to mid-level southwesterly winds in between them, particularly on 12–

14 July 2016 along the western border of the DACCIWA re- gion (arrows in Fig. 14b). This behaviour is associated with a weakening and northward shift in the AEJ (Fig. 8). It is con- ceivable that these westerly wind anomalies also helped to in- tensify coastal upwelling as shown in Fig. 3. Given the zonal distance between the two centres, both positive and negative vorticity signals are apparent in the Hovmöller plot shown in Fig. 9, although the negative one is only strong past 10

W.

Propagation of these two features is relatively slow, with about 7 m s

. While the signal in the northerlies at 850 hPa is somewhat patchy, the signal in the southerlies, created by the positive superposition of the wind disturbances associ- ated with the staggered northern and southern vortices, is co- herent and strong, particularly to the west of 10

W (Fig. 9), as also reflected in TCWV (Fig. 7). This has likely supported a deeper inland penetration and slight intensification of rain- fall (Fig. 5b). Given the somewhat unusual behaviour of this system, it is no surprise that there is no matching between the TDs and long-lived MCSs objectively identified during this period (Fig. 10). This propagating cyclonic–anticyclonic vortex couplet appears unrelated to any of the classical equa- torial waves, but the slow propagation speed and the oppos- ing circulation centres are consistent with the 6–9-day wave regime described by Diedhiou et al. (1999). To the best of our knowledge, the dynamical origin of such features is still somewhat unclear. In particular, the southern origin of the anticyclonic centre and its faster propagation seem unusual.

An interesting effect on the coastal region is that the south-

ern anticyclonic vortex (Feature H

in Fig. 14b) appears to

have brought with it dry air from the area of subsidence in